On performance of mutiuser diversity in SISO and MIMO wireless communication

Abstract

MIMO communication systems can offer much more advantages in terms of rate and reliability than SISO communication systems. In cellular wireless communication system, however, gains may be limited due to fading and interference. For downlink case, one potential solution is known as multiuser diversity, in which a packet scheduler improves throughput by exploiting the independence of the fading and interference statistics of different users. In this paper, the performance of downlink multiuser diversity with a large number of users for SISO and MIMO systems is considered. Through approximate analysis, it is found that in SISO systems, the multiuser diversity gain in terms of information capacity is A log(log K)/sup 1/ asymptotically, where K is the number of users, A is a constant. However in MIMO systems, if the same schedule strategy is used as in SISO systems, the multiuser diversity gain is upper bounded by Blog(log K/M)/sup 2/, where M is the number of antennas, B is a constant independent of M . When the number of transmit and receive antennas approaches infinity, multiuser diversity gain vanishes, which means that the total downlink information capacity doesn't increase with the increase of the number of users. It is also pointed out that the goal of optimizing a cellular system with multiple antennas is to maximize the total gain consisting of the multiuser diversity gain and the spatial multiplexing gain.